Skip to content

Add dynamic power for combinational cells#106

Draft
e-dworkin wants to merge 8 commits into
stineje:mainfrom
e-dworkin:feature/dynamic-power
Draft

Add dynamic power for combinational cells#106
e-dworkin wants to merge 8 commits into
stineje:mainfrom
e-dworkin:feature/dynamic-power

Conversation

@e-dworkin

Copy link
Copy Markdown

No description provided.

@e-dworkin e-dworkin force-pushed the feature/dynamic-power branch from 9d0dfec to d2aaecd Compare June 26, 2026 05:58
@infinitymdm

Copy link
Copy Markdown
Collaborator

Is Q/V pin capacitance required for the leakage and dynamic power work you're doing here? If it isn't absolutely required, I'd prefer it be submitted on a separate PR.

I've actually been meaning to add that for some time, just haven't had a moment where it was a higher priority than sequential stuff. That means a PR for your Q/V method would likely be merged sooner rather than later.

@e-dworkin

Copy link
Copy Markdown
Author

@infinitymdm I've made a new PR with just the Q/V method procedure. You're right, it isn't necessary for the dynamic power. I'm still cleaning up this PR for dynamic power and I'll open it for review hopefully next week.

As a side-note, I implemented the Q/V method procedure because the AC sweep procedure was giving me input capacitance values 10x what I was getting from Liberate using FreePDK45 models for a simple inverter chain. With the Q/V method I was getting more reasonable values. I plan to do more testing on the AC sweep specifically next week. If the problem isn't immediately obvious I'll mark it as an issue on GitHub and provide the relevant code so you can reproduce it yourself.

@e-dworkin

Copy link
Copy Markdown
Author

@rovinski @zpzhu99 FYI for dynamic power. Charge integration procedure for pin capacitance has been merged to main.

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

2 participants